Re: Bug#419197: libsane: xsane fails with kooka 2.6.20-1-686 works with 2.6.18-4-686
reassign 419197 linux-2.6
retitle 419197 CONFIG_USB_SUSPEND is experimental and breaks applications
severity 419197 important
thanks
Stuart Pook <debian47@stuart.es> wrote:
Hi,
> Xsane works fine when running Linux 2.6.18-4-686 however when I boot
> 2.6.20-1-686 I have the problem described by Michael Burschik. Xsane
> appears to sane but the scanner does not react and xsane receives a
> black scan. I have xsane 0.99+0.991-3.
This is due to CONFIG_USB_SUSPEND=y added in Linux 2.6.20, this
feature is still experimental and obviously it doesn't work that well
just yet.
Please disable this feature in the Debian kernel until this get fixed
somehow.
As to why scanimage works and the graphical frontends don't, either:
- the device node associated to the scanner changes everytime it's
put to sleep and woken up again
- the scanner doesn't like to be put to sleep for some reason
scanimage identifies the scanner and immediately proceeds with a scan,
leaving no time for the kernel to suspend the scanner. The graphical
frontends, on the other hand, identify the scanner(s) and then wait
for the user to start a scan. Once the scanner is identified, it's
identified using a SANE device string: backend:libusb:bus:dev. If the
identifier for the scanner changes (the device number on the bus),
like I suspect it might do, then we have a problem. One that we'll
have a hard time fixing, and I guess that we won't be the only ones.
That's only a somewhat educated guess, all that still needs to be
confirmed.
In any case, please Cc me on replies.
Thanks,
JB.
--
Julien BLACHE - Debian & GNU/Linux Developer - <jblache@debian.org>
Public key available on <http://www.jblache.org> - KeyID: F5D6 5169
GPG Fingerprint : 935A 79F1 C8B3 3521 FD62 7CC7 CD61 4FD7 F5D6 5169
Reply to: